Radar Operating Metrics and Network Throughput for Integrated Sensing and Communications in Millimeter-wave Urban Environments

Millimeter wave integrated sensing and communication (ISAC) systems are being researched for next-generation intelligent transportation systems. Here, radar and communication functionalities share a common spectrum and hardware resources in a time-multiplexed manner. The objective of the radar is to first scan the angular search space and detect and localize mobile users/targets in the presence of discrete clutter scatterers. Subsequently, this information is used to direct highly directional beams toward these mobile users for communication service. The choice of radar parameters such as the radar duty cycle and the corresponding beamwidth are critical for realizing high communication throughput. In this work, we use the stochastic geometry-based mathematical framework to analyze the radar operating metrics as a function of diverse radar, target, and clutter parameters and subsequently use these results to study the network throughput of the ISAC system. The results are validated through Monte Carlo simulations.

💡 Research Summary

This paper investigates the performance trade‑offs of millimeter‑wave (mmWave) integrated sensing and communication (ISAC) systems that share spectrum and hardware in a time‑multiplexed fashion, a configuration especially relevant for next‑generation intelligent transportation systems. The authors model a dual‑function monostatic radar/communication node located at the origin, which first scans the angular domain to detect and localize mobile users (targets) and then uses the acquired information to form highly directional communication beams.

A stochastic geometry framework is employed to capture the spatial randomness of both discrete clutter scatterers and mobile targets. Clutter points follow a homogeneous Poisson point process (PPP) with density ρ_c, and each scatterer’s radar cross‑section (RCS) σ_c is combined with ρ_c into a surface clutter coefficient σ_o = σ_c ρ_c. The target’s RCS follows a Swerling‑1 (exponential) distribution with mean σ_t̄. The received radar signal consists of the two‑way target return, the aggregate clutter return, and thermal noise N_p = k T_s B_W.

The paper derives closed‑form expressions for the probability of false alarm (P_fa) and probability of detection (P_d). Using the Gil‑Pelaez inversion theorem, the characteristic function of the aggregate clutter is obtained via the probability‑generating functional (PGFL) of the PPP. This leads to integral expressions for the cumulative distribution function (CDF) of clutter power, which are evaluated numerically to compute P_fa and P_d under various system parameters (transmit power P_tx, path‑loss exponent α, bandwidth B_W, duty cycle ξ, etc.).

Time‑multiplexing is modeled by splitting each ISAC cycle of duration T_total into a radar interval T_radar = ξ T_total and a communication interval T_comm = (1 − ξ) T_total. For a fixed dwell time per beam, the radar beamwidth Δψ is inversely proportional to ξ (Δψ = T_dwell Ω / (ξ T_total)), implying that a larger duty cycle yields narrower beams, higher antenna gain, and better detection probability, but also higher clutter‑induced false alarms and reduced communication time.

The average number of detected targets is expressed as β(ξ) = P_d ρ_t Ω (cτ/2), where ρ_t is the spatial density of users and cτ/2 is the range resolution. Assuming each detected user receives a fixed data rate D during the communication phase, the overall network throughput is γ = β(ξ) (1 − ξ) D. Thus γ depends on both the detection performance (which improves with larger ξ) and the available communication time (which shrinks with larger ξ).

Extensive Monte‑Carlo simulations validate the analytical results. Key findings include:

1. Transmit Power (P_tx): Higher P_tx raises both clutter power and target signal strength. Consequently, P_fa increases, but P_d improves more significantly, leading to a net increase in throughput γ.

2. Bandwidth (B_W): Increasing bandwidth reduces the clutter cell area A_r (since A_r ∝ 1/B_W), initially lowering P_fa. However, thermal noise grows linearly with B_W; beyond ≈100 MHz the noise dominates, causing P_fa to rise and γ to fall because the fixed data rate D does not scale with bandwidth in the model.

3. Radar Duty Cycle (ξ): P_fa grows monotonically with ξ due to higher antenna gain. Throughput γ exhibits a concave shape: it rises for small‑to‑moderate ξ (benefiting from higher P_d) and then declines sharply as the communication window shrinks. The optimal ξ lies around 0.5–0.7 depending on the path‑loss exponent α.

4. Surface Clutter Coefficient (σ_o): Larger σ_o dramatically increases P_fa, potentially reaching 100 % for very high clutter levels under low attenuation (α = 2). When the environment is highly lossy (α = 4), the impact of σ_o diminishes because both target and clutter returns are heavily attenuated.

The analysis demonstrates that ISAC system design must jointly optimize radar dwell time, beamwidth, transmit power, and bandwidth to balance detection reliability against communication capacity. The stochastic‑geometry‑based model provides a tractable way to predict performance in dense urban scenarios without exhaustive field measurements, and it highlights the importance of accounting for discrete clutter and non‑line‑of‑sight propagation. Future work is suggested to extend the framework to multi‑target, multi‑base‑station deployments, dynamic duty‑cycle scheduling, and experimental validation with real 802.11ad hardware.